550  PROF.  J.  W.  JUDD  ON  COMPOSITE  DYKES  IN  ARRAN.  [XoV.  I  893, 
simulate  the  appearance  of  globules  of  hyalite.1  In  order  to 
arrive  at  more  definite  conclusions  on  the  subject  I  have  submitted 
the  globules  to  the  following  microchemical  tests.  Treated  with 
caustic  potash,  the  globules  are  slowly  but  surely  attacked,  while 
the  glass  of  the  rock  is  unaffected  ;  this  slow  action  of  caustic 
potash  on  it  is  a  well-marked  character  of  hyalite.  After  long 
treatment  with  caustic  potash,  too,  these  globules  lost  their  perfect 
clearness  and  acquired  the  opalescence  of  ordinary  opals.  The 
globules  also  show  their  permeability  by  liquids  when  they  are 
treated  for  some  time  with  a  solution  of  fuchsine,  and  when,  after 
long  immersion  in  a  solution  of  sugar,  they  are  acted  upon  with 
sulphuric  acid.  I  think,  therefore,  that  there  can  be  little  doubt 
that  these  globules  are  really  composed  of  the  form  of  opal  known 
as  hyalite  or  4  Muller’s  glass.’ 
Although  opal  often  occurs  in  cavities  or  fissures  in  rocks,  this 
is  the  first  occasion,  I  believe,  in  which  it  has  been  shown  by 
microscopical  study  to  be  an  actual  rock-constituent.  It  will  be 
remembered,  however,  that  von  Lasauls  and  other  mineralogists 
have  argued  that  the  behaviour  of  the  groundmass  of  many  acid 
rocks  with  re-agents  points  to  the  existence  in  them  of  some  form 
of  opal.2  That  the  opal  in  the  rock  of  the  Cir  Mhor  dyke  is 
younger  than  the  porphyritie  crystals,  and  also  than  the  hornblende- 
microlites  of  the  groundmass,  is  perfectly  clear;  but  at  present  we 
have  no  trustworthy  data  to  show  whether  it  separated  before  or 
after  the  complete  consolidation  of  the  rock. 
In  addition  to  these  globules  of  hyalite  there  are  in  the  pitch- 
stone  of  the  Cir  Mhor  dyke  ordinary  spherulites,  showing  positive 
double-refraction.  These  are  somewhat  sparsely  scattered,  and 
are  at  once  distinguished  by  their  greater  opacity  (see  PL  XIX. 
fig.  1).  Thin  sections  viewed  with  high  powers  show  that  these 
spherulites  are  built  up  of  the  usual  forked  microlites  of  felspar, 
and  that  scattered  among  them  are  plates  of  tridymite.  The  nuclei 
of  many  of  these  spherulites  are  also  seen  with  high  magnifying- 
powers  to  be  formed  of  groups  of  overlapping  tridy mite-plates,  while 
similar  plates  are  found  scattered  among  the  felspar-microlites. 
The  rocks  of  the  Cir  Mhor  dyke  called  by  the  earlier  authors 
4  hornstone  ’  and  4  clay  stone -porphyry  ’  are  seen,  when  studied  under 
the  microscope,  to  represent  different  stages  in  the  devitrification 
of  the  4  pitchstone.’  In  the  rock  called  4  hornstone  ’  the  microscope 
shows  that  the  porphyritie  crystals  present  are  identical  with  those 
of  the  glassy  variety  of  the  rock,  but  that  some  changes  have  taken 
1  Prof.  Eosenbusch,  who  has  kindly  examined  a  section,  points  out  that  the 
most  delicate  tests  at  his  command  failed  to  show  any  difference  in  the  refrac¬ 
tive  index  between  these  clear  globules  and  the  surrounding  glass.  The 
refractive  index  of  obsidian  is  1'48  and  of  hyalite  1'45,  but  while  the  refractive 
index  of  the  glass  would  be  diminished  by  the  very  large  quantity  of  water 
shown  in  the  analysis,  that  of  the  hyalite  would  be  raised  by  the  numerous 
microscopic  and  ‘  ultra-microscopic  ’  needles  of  hornblende  which  it  contains. 
2  Mr.  Teall,  in  his  ‘British  Petrography’  (p.  309),  states  that  some  felsites 
of  St.  David’s  and  of  Arran  exhibit  spherulites  with  the  appearance  and  optical 
characters  of  hyalite. 
